Update app.py

app.py

@@ -57,7 +57,7 @@ def init_state() :
         st.session_state.repetion_penalty = 1
 
     if "rag_enabled" not in st.session_state :
-        st.session_state.rag_enabled = False
+        st.session_state.rag_enabled = True
 
     if "chat_bot" not in st.session_state :
         st.session_state.chat_bot = "Mixtral 8x7B v0.1"
@@ -80,7 +80,6 @@ def sidebar() :
         st.session_state.chat_bot = st.sidebar.radio('Seleziona Modello:', [key for key, value in CHAT_BOTS.items() ])
         st.session_state.temp = st.slider(label="Creatività", min_value=0.0, max_value=1.0, step=0.1, value=0.9)
         st.session_state.max_tokens = st.slider(label="Lunghezza Output", min_value = 64, max_value=2048, step= 32, value=512)
-        st.session_state.repetion_penalty = st.slider(label="Penalità Ripetizione", min_value=0., max_value=1., step=0.1, value=1. )
 
     with st.sidebar:
         retrieval_settings()
@@ -127,12 +126,16 @@ def stream_handler(chat_stream, placeholder) :
     tokens_per_second = total_tokens_processed // elapsed_time
     len_response = (len(prompt.split()) + len(full_response.split())) * 1.25
     col1, col2, col3 = st.columns(3)
-    
+
     with col1 :
-        st.write(f"**{tokens_per_second} token/secondi**")
-    
+        st.write(f"**{elapsed_time} secondi**")
+
     with col2 :
         st.write(f"**{int(len_response)} tokens generati**")
+
+    with col3 :
+        st.write(f"**{tokens_per_second} token/secondi**")
+    
         
     return full_response
 
@@ -161,4 +164,4 @@ if prompt := st.chat_input("Chatta con BonsiAI..."):
 
     st.session_state.history.append([prompt, full_response])
     st.session_state.messages.append({"role": "assistant", "content": full_response})
-    st.success('Generazione Completata', icon="✅")
+    st.success('Generazione Completata')